Nurse Aides’ Reports of Resident Abuse in Nursing Homes

Abstract

Verbal, physical, psychological, caregiving, medication, material, and sexual abuse perpetrated by staff are examined. Information came from nurse aides included in the Pennsylvania nurse aide registry. A total of 4,451 nurse aides returned a 46-item questionnaire (response rate = 64%). A 3-month frame of reference was used, and the questionnaire asked about abuse in the prior nursing home of employment. Nurse aide responses to the verbal abuse and psychological abuse items were higher than for the other categories of abuse examined. For example, 36% of nurse aides observed argumentative behavior with residents and 28% observed intimidation. Lower figures were reported for physical abuse (6% observed pushing, grabbing, or pinching), caregiving abuse (10% observed staff to threaten to stop taking care of a resident), medication abuse (19% observed inappropriate delays in medication administration), material exploitation (10% observed taking assets), and sexual abuse (1% observed unwelcome discussion of sexual activity). These findings clearly show that some types of resident abuse by staff are reported to be a common occurrence by nurse aides. This likely influences the health outcomes, quality of life, quality of care, and the safety of residents.

Keywords

nursing homes abuse nurse aides

Some nursing home research has estimated an annual rate of abusive events of 20.7 per 1,000 residents (Jogerst, Daly, Dawson, Peek-Asa, & Schmuch, 2006). The long-term care community in the United States includes an estimated 17,000 nursing homes (National Center for Health Statistics, 2004b). Currently, these nursing homes provide care and treatment for more than 3.6 million chronically ill and disabled elders per year (Centers for Medicare & Medicaid Services [CMS], 2009). Thus, the magnitude of abuse of elders in nursing homes may be substantial. In the research presented here, we present reports of the magnitude of resident abuse from a large sample of nurse aides.

Care and treatment in nursing homes is provided primarily by nurse aides (National Center for Health Statistics, 2004a). Specifically, nurse aides are estimated to provide approximately 80% of resident care (Brannon, 1992). Thus, they are in a unique position relative to nursing home residents, and their reports of resident abuse were used in this research. Definitions of elder abuse are varied. A broad definition of elder abuse, coming from the American Medical Association (AMA), was used in this research: “An act of commission or omission that results in harm or threatened harm to the health or welfare of an older adult” (Stiles, Koren, & Walsh, 2002, p. 34). Moreover, based on a review of the literature, interviews with directors of nursing and nurse aides, seven categories of elder abuse were examined in this research: verbal, physical, psychological, caregiving, medication, material, and sexual abuse (definitions are given below).

Clearly, abuse may have an extremely important impact on the daily lives of residents, having negative implications for both morbidity and mortality. Elder abuse, for example, is associated with adverse health consequences including depression (Lachs, Williams, O’Brien, & Pillemer, 2002). It certainly can also influence quality of life. Furthermore, as some have noted (Lachs et al., 2002), resident abuse is an issue of neglect, in addition to an issue of resident safety.

Existing Literature

The National Research Council (2003) issued a report summarizing the current state of scientific knowledge in the area of elder abuse, noting a variety of fundamental deficits. Among the issues raised by the report were inconsistent definitions of elder abuse, unclear and inadequate measures, and incomplete professional accounts. The council also recommended a research agenda to address these concerns, including more descriptive measurement in institutional settings. The research we present is a descriptive account of abuse in nursing homes.

Lachs and Pillemer (2004) stated, “There is almost no scientifically credible empirical research about abuse in institutions” (p. 1264). This was verified in a review of the literature that identified only five empirical studies addressing abuse in U.S. nursing homes (Lindbloom, Brandt, Hough, & Meadows, 2007). However, it is recognized that elders in nursing homes may be particularly vulnerable to abuse and neglect. This vulnerability stems from the cognitive impairment, behavioral abnormalities, and/or physical limitations of nursing home residents which have been reported as risk factors for abuse and neglect (Dyer, Connoly, & McFeeley, 2002).

Some research has attempted to substantiate incidence and prevalence rates of abuse. The National Elder Abuse Incidence Study (1998) identified that 50,000 cases of abuse were reported in nursing homes (49% were neglect, 35% emotional abuse, 30% physical abuse, 26% financial, 4% abandonment, and 1% sexual abuse). Ombudsman reported that 10% of nursing home complaints (i.e., approximately 20,000 complaints per year) were for abuse, gross neglect, or exploitation (U.S. Department of Health and Human Services, Administration on Aging, 2005).

A landmark study by Pillemer and Moore (1989) identified physical abuse, verbal abuse, and neglect to be common types of abuse in nursing homes. This research used responses from 577 staff (mostly nurse aides) in 31 facilities and identified 10% of these staff physically abused residents, while 40% of staff admitted to psychological resident abuse (Pillemer & Moore, 1989). The authors noted that verbal and physical abuse may be “basic features of nursing home life” (Pillemer & Moore, 1989, p. 314).

Allen, Kellett, and Gruman (2003), relying on the state’s Ombudsmen data, found that 47% of Connecticut’s nursing homes had one or more accounts of abuse, with physical abuse being reported more than other types. A more recent study by Jogerst and associates (2006) found an annual rate of abusive events of 20.7 per 1,000 nursing home residents, with an annual rate of 18.4 per 1,000 residents reported to authorities (i.e., Department of Inspections and Appeals). However, this study was limited to nursing homes in Iowa (N = 369) and outside inspection substantiated only 29% of cases (Jogerst et al., 2006). Other research has examined specific types of abuse, including financial and material exploitation. One study found 1.5% of staff reporting that they had stolen items from residents (Harris & Benson, 1999). Staff believed that 20% of their coworkers had also stolen items from residents (Harris & Benson, 1999).

These reported rates of abuse are likely underestimates. An estimated 84% of elder abuse cases go unreported (or 5 of 6 cases not reported; National Center on Elder Abuse, 1998). In 44 states and the District of Columbia, health care professionals, law enforcement, and clergy are mandatory reporters who are required to report allegations of abuse (Daly, Jogerst, Brining, & Dawson, 2003). Yet physicians report only 2% of all elder abuse and neglect cases (Rosenblatt & Kyung-Hwan, 1996). Kennedy (2005) found that despite 31% of physicians they surveyed admitting to having seen patients who presented with elder abuse and neglect, 94% did not report the abuse.

Pennsylvania nursing home administrators and Ombudsmen estimated that about 60% of institutional elder abuse cases were unreported (Peduzzi, Watzlaf, Rohrer, & Rubinstein, 1997). One reason for this is that many residents are unable to report abuse or neglect or they are fearful that reporting may lead to retaliation or otherwise negatively affect their lives (Compton, Flanagan, & Gregg, 1997). Other barriers to reporting include the lack of awareness of what abuse is and insufficient understanding of proper reporting procedures (Swagerty, Takahashi, & Evans, 1999).

In addition to the high rates of underreporting, elder abuse and neglect becomes difficult to study due to varying definitions, reporting protocols, and investigation standards across the states. As Roby and Sullivan (2000) discuss, variations in state laws, including 28 different terms for elder abuse and neglect, have made it difficult to compare empirical data from different states. In addition, of all licensing and recertification statutes of nursing homes across the states, only 27% had elderly abuse reporting and investigation information (Daly & Jogerst, 2007). Of these states, language varied considerably regarding mandatory reporting, investigation processes, and definitions of types of facilities and types of abuse (Daly & Jogerst, 2007).

Significance of Elder Abuse

There is a growing awareness that the causes of death of nursing home residents are often unknown. Deaths of individuals with long-standing, chronic illness with multiple comorbidities often go uninvestigated and can be incorrectly identified. However, evidence suggests that abuse and neglect of elderly persons is associated with increased mortality rates (Ortmann, Fechner, Bajanowski, & Brinkman, 2001). A study of 2,400 deaths in Arkansas nursing homes found 50 cases of suspected abuse or neglect (Ortmann et al., 2001). It was found that the risk of death for elder abuse and neglect victims are three times higher than for nonvictims (American Medical Association White Paper on Elderly Health, 1990).

The health implications of abuse also include an association with morbidity (Watts & Zimmerman, 2002). Although not specifically from a nursing home context, morbidity associated with elder abuse includes fractures, depression, dementia, and malnutrition (Levine, 2003; Dyer, Pavlik, Murphy, & Hyman, 2000). The direct medical costs associated with injuries from physical abuse are estimated to add more than US$5.3 billion to the nation’s annual health expenditures (Mouton et al., 2004).

This Research

The studies reviewed above highlight the need for additional research on resident abuse in nursing homes. Yet several barriers previously noted (such as underreporting, varying definitions, and differences in state regulations) make elder abuse in nursing homes difficult to study. The approach used here was to examine reports of the magnitude of resident abuse from a large sample of nurse aides.

For some nurse aides, a possible psychological barrier of reporting abuse in their current nursing home of employment may exist. For example, resident abuse may be occurring (or has previously occurred) and they may not have intervened. To overcome this, nurse aides were asked about abuse in their prior nursing home of employment.

The perpetrators of abuse can also vary and include family, caregivers (i.e., staff), and even other residents (Shinoda-Tagawa, Leonard, & Pontikas, 2004). In this research, we focus on resident abuse perpetrated from staff caregivers.

Method

Source of Data

Information on resident abuse came from a survey of nurse aides (also often called Certified Nursing Assistants [CNAs]). Nurse aides were identified because of their inclusion in the Pennsylvania nurse aide registry. All states are required by Centers for Medicare & Medicaid Services (CMS) to maintain a nurse aide registry (Elvidge & Buechlein, 1992). The purpose of these state registries is to “ensure that nurse aides have education, practical knowledge, and skills needed to care for residents of facilities participating in the Medicare and Medicaid programs” (Federal Register, 1991, p. 48880). This also includes the exclusion of nurse aides identified as abusing elders. Federal (e.g., 42 CFR 483.75(e)(5)) and State regulations (e.g., for Pennsylvania Code § 1181.531) require nursing homes to check this nurse aide registry, verify each nurse aide’s certification status, and only hire nurse aides that are included in the registry. For example, in Pennsylvania, instruction to facilities is as follows: “A nurse aide who is not enrolled or in good standing on the registry may not be employed in a nursing care facility that receives Medicare or Medicaid reimbursement” (Pennsylvania Department of Health, 2010).

A random sample of approximately 20% (N = 7,000) of nurse aides in the Pennsylvania registry was used. This sample size was determined based on the resources available for the research, likely response rates (using information from prior surveys we conducted with nurse aides; e.g., Castle, Wagner, Ferguson, & Handler, in press), and power calculations (using published information from the studies reviewed above). The survey was sent along with a cover letter describing the study and a prepaid return envelope. Follow-up reminder post cards were mailed 2 and 4 weeks after the survey mailing. A total of 4,451 nurse aides returned the questionnaire (giving a response rate of 64%).

Abuse Questionnaire Development

The abuse questionnaire items are discussed in the following section. First, the development process used for this questionnaire is described. Development consisted of four steps: determining which areas of abuse to include, formulating questions, refining questions, and cognitive testing of the draft instrument.

The first step in developing the questionnaire was to determine which areas of abuse to investigate. This was necessary because the scope of resident abuse in nursing homes is wide ranging (see a review by Daly & Jogerst, 2007). The emphasis in this research was on verbal, physical, psychological, caregiving, medication, material, and sexual abuse. Definitions of each are provided below.

This emphasis was determined based on a review of the literature, interviews with four directors of nursing, and interviews with 10 nurse aides. The nursing homes we used in this development process (and subsequent development steps discussed below) were all located in Pennsylvania, and nurse aides (i.e., CNAs) were selected based on recommendations from the directors of nursing of these facilities. Thus, this represented a sample of convenience.

A questionnaire was developed for this research because few existing abuse instruments were found to be extensively used or developed with long-term care settings in mind, and few previous surveys have used nurse aides as respondents. An exception is the survey used by Pillemer and Moore (1989); however, this prior survey did not include all of the categories of abuse in which we were interested. Items from this prior survey were included in our survey development process, described below.

Nurse aides typically do not have high levels of education, which may limit understanding of questions included in existing instruments (Castle, 2008a). To help construct items for the questionnaire, interviews with an additional 37 nurse aides from 10 of the Pennsylvania nursing homes (discussed above) were used. Nurse aide interviews were not conducted in the nursing homes. The interviews used an open-ended question format, and nurse aides were asked their opinions on potential abuse items and for opinions on wording questions in general. Based on these interviews with nurse aides, the pool of items was refined. That is, items were written to be relevant to nurse aides (i.e., face validity), to be relevant to the nursing home context (i.e., content validity), and to be easily understood. A Flesch-Kinkaid Scale score of 9 or lower was achieved for all questions (Flesch, 1948). The Flesch-Kinkaid Scale (Streiner & Norman, 1995) is an index of readability based on the average number of syllables per word and average number of words per sentence. This implies that for the abuse questionnaire a respondent with a ninth-grade education should correctly understand all the questions being asked.

The pool of items (N = 155) was also reduced by eliminating duplicate questions, very similar questions, and by choosing the most relevant questions (again, based on feedback from nurse aides). After this process, a total pool of 52 questions remained.

The final step in developing this questionnaire consisted of cognitive testing with an additional 15 nurse aides. These nurse aides were selected in the same way as the 37 selected for interviews (above) and from the same nursing homes. Cognitive testing is commonly used in conjunction with questionnaire development. Briefly, this consists of face-to-face interviews to evaluate whether respondents fully understand words, phrases, and concepts (R. E. Levine, Fowler, & Brown, 2005). For example, this approach helps determine whether potential respondents answer a question in the way it is intended to be answered in the survey. Cognitive testing can also be used to refine the wording of items (Levine et al., 2005). In this case, the process resulted in six questions being removed from the pilot instrument and minor changes to the wording of a further four questions.

Abuse Questionnaire Content

On completion of this development process, the questionnaire had 46 items. These items included subscales addressing verbal, physical, psychological, caregiving, medication, material, and sexual abuse. Each subscale also included an open-ended question, asking for any additional comments. In addition, other items addressed demographic characteristics and characteristics of prior employment. These subscales are described further.

Verbal abuse is defined as “intentional infliction of anguish, pain or distress through verbal” acts (Clarke & Pierson, 1999, p. 635). Items specifically ask questions about yelling, insulting remarks, and cursing. For example, one item asks “have you heard a staff member curse at a resident?”

Physical abuse is defined as “acts done with the intention of causing physical pain or injury” (Lachs & Pillemer, 2004, p. 1264). Items specifically ask questions on pushing, grabbing, or pinching, pulling hair or kicking, and other physical violence. For example, one item asks “have you seen a staff member kick a resident?”

Psychological abuse is defined as “acts done with the intention of causing emotional pain or injury” (Lachs & Pillemer, 2004, p. 1264). Items specifically ask questions on intimidation, aggressive behavior, threatening remarks, and critical remarks. For example, one item asks “have you seen a staff member threaten a resident?”

Caregiving abuse is defined as “failure to provide goods and services necessary to avoid physical harm, mental anguish or mental illness” (Federal Register, 1991, p. 48887). Items specifically ask questions on withholding care, or threatening to withhold food and water. For example, one item asks “have you seen a staff member threaten to stop taking care of a resident?”

Medication abuse is defined as elders “purposely deprived of their correct medication or given inappropriate medication” (Chambers, 1999, p. 80). Items specifically ask questions on withholding medications, excessive medications, and access to medications. For example, one item asks “have you seen a staff member purposefully withhold medication from a resident?”

Material exploitation is defined as “the improper use of an older person’s assets” (Rabiner, O’Keeffe, & Brown, 2006, p. 51). Items specifically ask questions on taking possessions and taking assets. For example, one item asks “have you seen a staff member take any belongings of a resident?”

Sexual abuse is defined as “non-consenting sexual contact of any kind” (National Center on Elder Abuse, 1998, p. 1). Items specifically ask questions on unwelcome sexualized kissing, unwelcome fondling, unwelcome discussion of sexual activity, exposure of private body parts to embarrass, oral-genital contact, digital penetration, and vaginal rape. For example, one item asks “have you seen any staff involved in unwelcome discussion of sexual activity with a resident?”

Other items in the questionnaire address nurse aide demographic characteristics (e.g., race, age, gender, tenure) and characteristics of employment (i.e., shift, number of residents cared for, primary unit). When abuse was identified, the number of cases was also requested (which is used for calculating approximate prevalence statistics).

Abuse Questionnaire Format

Three questionnaire formatting items of importance were the issues of observing abuse, the reference period, and employment in a prior facility. Many forms of abuse can be perpetrated quickly, making observation less likely. Therefore, questions were asked using multiple formats: (a) observed or have evidence that this happened; (b) the resident told you this happened; (c) someone other than the resident told you this happened; and, (d) you suspect that this happened.

Most items used a stem asking for information in the past 3 months of employment in a prior facility. This time period of 3 months was chosen based on our interviews with nurse aides. That is, nurse aides believed they could provide accurate response if this time frame was used. However, it is noted that the use of this time frame has limitations including recall by nurse aides and in this case the limitation of reducing the pool of respondents because they were not employed for this time duration (i.e., turnover of nurse aides in many facilities is high).

Nurse aides were asked about abuse in their prior place of employment. Response options were used to determine if the prior employer was a nursing home. As we note above, this may help alleviate a possible psychological barrier of reporting abuse in their current place of employment. Given the rapid turnover of nurse aides in nursing homes the time period for recall for many nurse aides is not long (Castle, 2008b). Thus, all abuse items in the questionnaire used the stem “During the last 3 months, in your last place of employment. . . .” Nevertheless, as we describe below several nurse aides are excluded from the sample by using this approach and addressing both the time period of interest and prior employer may tax some respondents.

Analyses

Descriptive analyses are presented, consisting of the percentages and means for the nurse aide sample and characteristics of the nursing homes in which they were employed. In addition, comparisons for respondent characteristics with characteristics of national averages were conducted.

Descriptive analyses are also presented for questionnaire items. That is, for each question the percentage of nurse aides responding positively (i.e., abuse occurred) and the number of respondents is given. The results are presented in this way for each of the response formats used (observed or have evidence that this happened; the resident told you this happened; someone other than the resident told you this happened; you suspect that this happened).

Prevalence is a measure of the number of observations of an event that occur in a population during a specified time period (Last, 1995). In this research, an approximate prevalence rate was calculated by using information on the number of individual residents identified as abused by staff assigned to the nurse aide divided by the number of assigned residents cared for in 3 months. The “observed or have evidence” category was used in making these calculations, and the figures are given as yearly rates. However, we note that this is presented as an approximate rate, as the information comes from self-reports by nurse aides.

Results

Of the 4,451 surveys returned some exclusions were made prior to analyses. Specifically, nurse aides with less than 3 months experience in the prior facility, did not work in a nursing home, had not primarily worked directly with residents, or were terminated from their previous position were excluded. This reduced the analytic sample to 3,433 nurse aides (giving an analytic response rate of 49%). We note that this response rate is a conservative estimate. Nurse aides who were not recently employed in a nursing home or had changed address could not be excluded from receiving a survey and are included in the denominator when calculating the response rate. However, these nurse aides could not complete the survey and be included in the numerator of the calculated response rate. Of the nurse aides returning the questionnaire, most answered all of the items.

Sample Characteristics

Table 1 presents descriptive statistics of the nurse aide sample, along with characteristics of the nursing homes in which they worked. Most nurse aide characteristics of the sample were not significantly different from equivalent characteristics recorded in the 2004 National Nursing Assistant Survey (NNAS; National Center for Health Statistics, 2004a). The NNAS was conducted as a supplement to the 2004 National Nursing Home Survey (NNHS; National Center for Health Statistics, 2004b). The NNHS is a nationally representative study of nursing homes conducted by the National Center for Health Statistics (http://www.cdc.gov/nchs/nnhs.htm). Nurse aides were most likely to be about 33 years old and female. Nurse aides were not asked to identify the prior nursing home in which they worked but were asked some basic characteristics (such as ownership and size). The average organizational size, for-profit status, and chain membership identified by nurse aides of their former employer were not significantly different from the averages reported in the NNHS.

Table 1.

Characteristics of Nurse Aides and Nursing Homes

Characteristic	%	M (SD)
Personal characteristics of nurse aides (N = 3,433)
Gender (female)	91
Age (years)		32.5 (8.1)
Race (minority)	44
Marital status (single)	25
Highest level of education (high school)	82
Tenure in current facility (years)		1.1 (3.1)
Facility characteristics of nursing homes^a
Facility size (number of beds)		125.1 (72.2)
For-profit ownership	55
Chain membership	43
Average occupancy	90

Taken from nurse aide responses.

Nurse Aides’ Opinions of Abuse

Nurse aide responses to the verbal abuse and psychological abuse items were higher than for the other categories of abuse examined (shown in Table 2). For example, 36% of nurse aides observed argumentative behavior with residents and 28% observed intimidation. Lower figures were reported for physical abuse (6% observed pushing, grabbing, or pinching), caregiving abuse (10% observed staff to threaten to stop taking care of a resident), medication abuse (19% observed inappropriate delays in medication administration), material exploitation (10% observed taking assets), and sexual abuse (1% observed unwelcome discussion of sexual activity).

Table 2.

Nurse Aides’ Opinions of Staff Abuse in Nursing Homes

	Observed or have evidence		The resident told you		Someone other than resident told you		You suspect
Category and item	%	N	%	N	%	N	%	N
Verbal abuse from staff
Argumentative with resident	36	1,226	30	1,025	20	678	36	1,225
Humiliating remarks	29	1,004	38	1,311	13	439	32	1,099
Yelling	27	924	42	1,457	7	254	39	1,345
Cursing	27	927	36	1,222	8	267	29	1,005
Insulting remarks	25	866	36	1,230	4	145	27	930
Physical abuse from staff
Pushing, grabbing, or pinching	6	216	6	212	6	216	12	423
Deliberately hurt resident	4	154	22	767	10	347	13	455
Hit or throw things at resident	4	146	15	510	4	154	9	320
Pulling hair or kicking	3	110	4	133	9	312	4	121
Other physical violence	3	86	7	252	7	237	3	120
Psychological abuse from staff
Intimidation	28	974	32	1,115	14	478	33	1,124
Threatening remarks	26	899	35	1,186	17	576	37	1,267
Critical remarks	21	732	54	1,865	15	520	48	1,632
Aggressive behavior	20	684	22	765	14	489	25	853
Caregiving abuse from staff
Threaten to stop taking care of resident	10	327	13	432	5	156	16	536
Deliberately withheld water	5	173	8	276	8	276	7	235
Deliberately withheld food	4	126	9	321	4	128	10	345
Medication abuse from staff
Inappropriately delayed giving medication	19	659	39	1,344	17	569	39	1,344
Denied access to medications	6	216	18	634	10	357	12	398
Given excessive medication	6	210	23	785	12	423	18	629
Material exploitation from staff
Taking assets	10	338	19	656	4	145	14	469
Destroyed belongings	7	230	10	344	17	579	8	259
Taking possessions	6	219	26	890	6	217	22	767
Sign checks without permission	3	92	7	232	4	127	4	136
Tamper with savings	2	62	13	438	6	189	6	217
Sexual abuse from staff
Exposure of private body parts to embarrass	2	69	6	216	5	176	13	431
Unwelcome fondling	1	28	4	126	4	147	7	241
Unwelcome discussion of sexual activity	1	36	4	143	7	257	8	266
Oral-genital contact	<1	12	<1	15	4	148	1	22
Digital penetration	<1	8	1	18	4	135	<1	15
Vaginal rape	<1	7	<1	12	1	37	<1	9

Note: Sample size = 3,433 nurse aides. Items in each category are listed from highest to lowest percentage in the “Observed or have evidence” category.

In general, higher levels of abuse were reported in the “resident told you” and “you suspect” categories. For example, with verbal abuse 42% of nurse aides were told by a resident about yelling and 39% suspected verbal abuse. In addition, 22% of nurse aides were told by a resident about staff deliberately hurting them, and 13% suspected this physical abuse. With respect to caregiving abuse, 13% of nurse aides were told by a resident about staff threatening to stop taking care of them, and 16% suspected this abuse.

Estimated Prevalence Rates

Prevalence rates using the observed or have evidence category and given as yearly rates are presented in Table 3. Following the nurse aides opinions of abuse, the highest prevalence rates were for verbal abuse (32% rate for argumentative behavior with residents) and psychological abuse (18% rate for intimidation). Generally, lower prevalence rates were reported for physical abuse, caregiving abuse, medication abuse, material exploitation, and sexual abuse.

Table 3.

Estimated Prevalence Rates of Staff Abuse in Nursing Homes

Category and item	Prevalence per 1,000 residents per year^a (%)
Verbal abuse from staff
Argumentative with resident	32
Humiliating remarks	25
Yelling	22
Insulting remarks	21
Cursing	21
Physical abuse from staff
Pushing, grabbing, or pinching	3
Deliberately hurt resident	2
Hit or throw things at resident	2
Pulling hair or kicking	1
Other physical violence	1
Psychological abuse from staff
Intimidation	18
Aggressive behavior	15
Critical remarks	14
Threatening remarks	11
Caregiving abuse from staff
Threaten to stop taking care of resident	8
Deliberately withheld food	1
Deliberately withheld water	1
Medication abuse from staff
Inappropriately delayed giving medication	13
Denied access to medications	2
Given excessive medication	2
Material exploitation from staff
Destroyed belongings	4
Taking assets	2
Taking possessions	1
Sign checks without permission	<1
Tamper with savings	<1
Sexual abuse from staff
Unwelcome fondling	<1
Unwelcome discussion of sexual activity	<1
Exposure of private body parts to embarrass	<1
Oral-genital contact	<1
Digital penetration	<1
Vaginal rape	<1

Note: Prevalence rate was calculated by using information on the number of individual residents identified as abused assigned to the nurse aide by staff divided by the number of assigned residents cared for in 3 months. Figures are given as yearly rates. Items in each category are listed from highest to lowest percentage.

Calculated using responses from 3,433 nurse aide.

Discussion

In 2005, US$5 million of federal funding was appropriated to address elder abuse and neglect in the United States. This level of funding pales in comparison to monies appropriated for other types of abuse. Child abuse received nearly US$7 billion to be used for services, research, training, and prevention, and the Violence Against Women Act provided more than US$500 million (Quinn & Zielke, 2005). This disparity in funding may be fuelled by the perception that elder abuse is uncommon. Our findings, for elder abuse in nursing homes, show that this may not be the case. Some types of elder abuse would appear to be highly prevalent.

Extrapolating the prevalence rates identified in this study and assuming approximately 3.6 million elders use a nursing home each year, 792,000 nursing home residents experienced verbal abuse (i.e., yelling) from staff and 648,000 experienced psychological abuse (i.e., intimidation). Although, of course, these figures need to be interpreted taking into consideration that they come from the opinions of nurse aides and no severity is linked to the abuse. That is, what nurse aides consider to be “yelling” and “intimidation” need to be clarified to further interpret these findings.

Moreover, these findings serve to highlight additional research that is needed to further understand abuse in nursing homes. Some clarity is also needed regarding which residents experience abuse. Many nursing home residents are now elders with short stays (CMS, 2009). Prior research has identified more impaired elders to be at greater risk of abuse (Dyer, Connoly, & McFeeley, 2002). Even with substantial downward adjustment, our findings show some types of abuse to be common in nursing homes. Also, abuse can create an environment in which other none abused residents can feel unsafe or threatened. That is, the implications of abuse go beyond the abuse victims.

Practice Implications

Some types of abuse identified are not readily amenable to interventions frequently used in community settings such as use of police services (e.g., insulting remarks). However, this should not relegate the potential importance of abuse. Insulting remarks could potentially lead to a cascade of events in elders, resulting in depression and even death. In the nursing home context, abuse such as insulting remarks may be extremely important given the circumscribed locus of control that exists for many residents. Many impaired residents have few interactions with friends, family, or other elders. For many, interactions with nursing home staff caregivers are the most important social interactions of their daily lives. In this context, no cases of abuse from staff caregivers can be considered benign.

Options to prevent abuse in nursing homes are needed (other than options used in the community). For example, Pillemer and Hudson (1993) present a training program for nurse aides based on materials from The Coalition of Advocates for the Rights of the Infirm Elderly (CARIE).

For this research, nurse aide respondents were identified in the Pennsylvania nurse aide registry. One function of this registry is to prevent known abusers of elders working in nursing homes. Similar registries exist in other states. However, for such registries to work as designed, abusers need to be first identified. With chronic underreporting of elder abuse, nurse aide registries may be failing in this essential function. Further interventions may be needed to improve reporting in this area. However, for the nursing home this raises a series of conflicting issues. Resident care should be paramount; however, identifying resident abuse has legal implications, influences staffing levels, and imperils careers of staff.

The scope of abuse examined in this research was wider than that of most other studies. For example, items were included addressing medications abuse. Given that many elders on average take nine or more different medications on a daily basis (the CMS definition for polypharmacy), the potential for abuse in this area is substantial (Doshi, Shaffer, & Briesacher, 2005). Findings from this research suggest that medication administration is less than optimal. Medication abuse could equate to adverse outcomes for many elders.

Moreover, from open-ended comments (not reported), many nurse aides suspect abuse with specific medications, especially those with some black-market value such as pain medications. However, given the highly regulated dispensing of these medications more detail on these comments would seem appropriate.

Following other limited research in this area (e.g., Teaster, Lawrence, & Cecil, 2007), the levels of sexual abuse identified were relatively low. In one respect, this is clearly positive. However, in another respect, the levels identified are not zero. Some nursing home residents are still at risk of sexual abuse.

Limitations and Suggestions for Future Research

As noted above, little is known about elder abuse in long-term care settings. As part of our literature review, a few international studies in long-term care settings were also identified. These studies used relatively small sample sizes, and include those by Buzgova and Ivanova (2009) in the Czech Republic (N = 511), Saveman, Astom, Bucht, and Norberg (1999) in Sweden (N = 499), Weatherall (2001) in New Zealand (N = 26), and Goergen (2004) in Germany (N = 251). Of interest, most of these studies also note how little we know about elder abuse. Our findings thus clearly give some information to an understudied problem. However, the findings presented are for nursing homes. More information is needed for elder abuse in other long-term care settings such as assisted living facilities (which in the United States now outnumber nursing homes).

We also note above that open-ended questions were used for each subscale. This resulted in several hundred written comments for each subscale. Qualitative analyses of these comments may also provide valuable information on nurse aides’ opinions of abuse in nursing homes.

It is possible to directly ask some residents about abuse. However, limitations to this approach are mentioned above, and such interviews are expensive. A mail survey of nurse aides is cost effective, and the anonymity of the survey process could have improved the reliability of the information obtained. Nevertheless, a concern is the willingness to report abuse. Prior research has shown nurse aides are willing to report abuse (Pillemer & Moore, 1989). The findings of our study also show that it is possible to solicit responses on abuse from nurse aides. Still, a limitation of this study is the unknown accuracy of the abuse reported. With no gold standard available, it is difficult to ascertain the accuracy of the responses received.

With respect to the responses received, it is likely that the findings we present are in some cases underestimates of abuse. That is, we were not able to determine the accuracy of the responses received; however, the sensitivity and subjectivity of the items were assessed. Cases of each type of abuse we examined were presented to 50 nurse aides (in additional analyses). In turn, nurse aides were asked to rate whether they would rate the case as an example of abuse and the severity of the abuse. Substantial agreement existed on sexual abuse, medication abuse, material exploitation, and physical abuse. However, much less agreement existed with respect to verbal, psychological, and caregiving abuse. In these cases, the threshold for abuse was higher. For example, nurse aides appeared somewhat indifferent to yelling and insulting remarks. Thus, we believe the responses received in this area may be underestimates—with the caveat that these items are subjective and no agreed-upon definition of abuse in these areas exists.

Material exploitation was examined in this research. Items such as tampering with savings had low reported rates. Nevertheless, with the recent economic downturn, levels of material exploitation may have changed. Moreover, few studies have examined this area of abuse (e.g., Choi, Kulick, & Mayer, 1999; MetLife, 2009; Rabiner, Brown, & O’Keeffe, 2004). The ability of residents to obtain and retain money may be more difficult because they are in an institutional setting. Money management may require help from others, giving the opportunity for exploitation. This may be a further area where what nurse aides consider to be material exploitation needs to be clarified to further interpret these findings. For example, taking a few dollars may not register as resident abuse from the nurse aides’ perspective. However, from the residents’ point of view, the taking of a few dollars may represent a considerable infraction.

To help frame this descriptive study, seven categories of abuse were used. However, standard definitions for abuse do not exist and other categories (and items within categories) could be used. For example, it could be argued that medication abuse is a form of caregiver neglect. In addition, nurse aides may interpret these definitions differently from residents and informal caregivers. Factor analyses will be used in further examining our data to help with additional scale refinement.

Nursing home residents are also at risk from other abusers, including family and even other residents, the questionnaire could be further developed by examining abuse from these other sources. This may be especially true for resident-to-resident abuse (Shinoda-Tagawa et al., 2004).

In addition, we note that, in general, nurse aides were especially concerned in the focus groups with lapses of caregiving for residents as issues of abuse (or even “criminal” in their words). This concern included well known clinical issues such as use of physical restraints and the incidence of pressure ulcers; however, it also included characteristics such as staffing levels. We echo this concern and believe the findings of this research should also be interpreted with these other lapses in caregiving as context.

Conclusion

Very little is known about resident abuse in nursing homes. With the proviso that this research used reports coming from nurse aides, the findings clearly show that some types of resident abuse by staff are reported to be quite prevalent. Verbal abuse and psychological abuse of residents by staff are reported as common. This may have an impact of the health of nursing home residents and likely influences their quality of life, quality of care, and safety. With the rising numbers of extremely vulnerable elders in long-term care facilities, it is important to further measure, understand, and eliminate abuse in these settings.

Footnotes

The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

This research was supported in part by a grant from the National Institute on Aging (NIA; 1R21AG028015-01: Testing survey methods for collecting data on elder mistreatment prevalence).

References

Allen

Kallett

Gruman

(2003). Elder abuse in Connecticut’s nursing homes. Journal of Elder Abuse & Neglect, 15(1), 19-42.

American Medical Association White Paper on Elderly Health. (1990). Report of the Council on Scientific Affairs. Archives of Internal Medicine, 150, 2459-2472.

Brannon

(1992). Toward second-generation nursing home research. The Gerontologist, 32, 293.

Buzgova

Ivanova

(2009). Elder abuse and mistreatment in residential settings. Nursing Ethics, 16(1), 110-126.

Castle

N. G.

(2008a). Nursing home caregiver staffing levels and quality of care: A literature review. Journal of Applied Gerontology, 27, 375-405.

Castle

N. G.

(2008b). State differences and facility differences in nursing home staff turnover. Journal of Applied Gerontology, 27, 609-630.

Castle

N. G.

Wagner

L. M.

Ferguson

J. C.

Handler

S. M.

(in press). Comparing the safety culture of nursing homes and hospitals. Journal of Applied Gerontology.

Centers for Medicare & Medicaid Services. (2009). Skilled nursing facilities non swing bed—Medicare national summary. Retrieved from www.cms.hhs.gov/MedicareFeeforSvcPartsAB/Downloads/NationalSum2007.pdf

Chambers

(1999). Potential for the abuse of medication for the elderly in residential and nursing homes in the UK. In Glendenning

Kingston

(Eds.), Elder abuse and neglect in residential settings (pp. 79-90). New York: Haworth.

10.

Choi

Kulick

Mayer

(1999). Financial exploitation of elders: Analysis of risk factors based on county adult protective services data. Journal of Elder Abuse and Neglect, 10, 362-369.

11.

Clarke

M. E.

Pierson

(1999). Management of elder abuse in the emergency department. Emergency Medical Clinics of North America, 17, 631-644.

12.

Compton

Flanagan

Gregg

(1997). Elder abuse in people with dementia in Northern Ireland: Prevalence and predictors in cases referred to psychiatry of old age service. International Journal of Geriatric Psychiatry, 12, 632-635.

13.

Daly

J. M.

Jogerst

G. J.

(2007). Nursing home abuse report and investigation legislation. Journal of Elder Abuse & Neglect, 19(3/4), 119-131.

14.

Daly

Jogerst

Brining

Dawson

(2003). Mandatory reporting: Relationship of APS statute language on state reported elder abuse. Journal of Elder Abuse and Neglect, 15(2), 1-21.

15.

Doshi

J. A.

Shaffer

Briesacher

B. A.

(2005). National estimates of medication use in nursing homes: Findings from the 1997 Medicare Current Beneficiary Survey and the 1996 Medical Expenditure Survey. Journal of the American Geriatrics Society, 53, 438-443.

16.

Dyer

Connoly

McFeeley

(2002). The clinical and medical forensics of elder abuse and neglect. In Bonnie

Wallace

(Eds.), Elder abuse: Abuse, neglect, and exploitation in an aging America (pp. 339-381). Washington, DC: National Academy Press.

17.

Dyer

Pavlik

Murphy

Hyman

(2000). The high prevalence of depression and dementia in elder abuse and neglect. Journal of the American Geriatrics Society, 48, 205-208.

18.

Elvidge

Buechlein

(1992). Nurse aide registry rules. Contemporary Long-Term Care, 15(10), 72.

19.

Federal Register. (1991, September 26). Medicare and Medicaid programs; nurse aide training and competency evaluation programs—HCFA. Final Rule, 56(187), 48880-48922.

20.

Flesch

(1948). A new readability yardstick. Journal of Applied Psychology, 32, 221-233.

21.

Goergen

(2004). A multi-method study on elder abuse and neglect in nursing homes. Journal of Adult Protection, 6(3), 15-25.

22.

Harris

D. K.

Benson

M. L.

(1999). Theft in nursing homes: An overlooked form of elder abuse. Journal of Elder Abuse & Neglect, 11(3), 73-90.

23.

Jogerst

G. J.

Daly

J. M.

Dawson

J. D.

Peek-Asa

Schmuch

(2006). Iowa nursing home characteristics associated with reported abuse. Journal of American Medical Directors Association, 7, 203-207.

24.

Kennedy

(2005). Elder abuse and neglect: The experience, knowledge, and attitudes of primary care physicians. Family Medicine, 37, 481-485.

25.

Lachs

M. S.

Williams

C. S.

O’Brien

Pillemer

(2002). Adult protective service home use and nursing home placement. The Gerontologist, 42, 734-739.

26.

Lachs

M. S.

Pillemer

(2004). Elder abuse. Lancet, 364, 1263-1272.

27.

Last

J. M.

(1995). A dictionary of epidemiology. New York, NY: Oxford University Press.

28.

Levine

R. E.

Fowler

F. J.

Brown

J. A.

(2005). Role of cognitive testing in the development of the CAHPS hospital survey. Health Services Research, 40(Part II), 2037-2056.

29.

Levine

(2003). Elder neglect and abuse: A primer for primary care physicians. Geriatrics, 58, 37-44.

30.

Lindbloom

E. J.

Brandt

Hough

L. D.

Meadows

S. E.

(2007). Elder mistreatment in the nursing home: A systematic review. Journal of American Medical Directors Association, 8, 610-616.

31.

MetLife. (2009). Broken trust: Elders, family and finances. Westport, CT: MetLife Mature Market Institute.

32.

Mouton

Rodabough

Rovi

Hunt

Talamantes

Brzyski

Burge

(2004). Prevalence and 3 year incidence of abuse among postmenopausal women. American Journal of Public Health, 94, 605-612.

33.

National Center for Health Statistics. (2004a). National Nursing Assistant Survey. Hyattsville, MD: Public Health Service.

34.

National Center for Health Statistics. (2004b). National Nursing Home Survey. Hyattsville, MD: Public Health Service.

35.

National Center on Elder Abuse. (1998). National Elder Abuse Incidence Study. Washington, DC: American Public Human Services Association.

36.

Ortmann

Fechner

Bajanowski

Brinkman

(2001). Fatal neglect of the elderly. International Journal of Legal Medicine, 114, 191-193.

37.

Peduzzi

Watzlaf

Rohrer

Rubinstein

(1997). A survey of nursing home administrators’ and ombudsmen’s perception of elderly abuse in Pennsylvania. Top Health Information Management, 18(1), 68-76.

38.

Pennsylvania Department of Health. (2010). Nurse aide registry. Retrieved from http://www.portal.state.pa.us/portal/server.pt?open=514&objID=558490&mode=2

39.

Pillemer

Hudson

(1993). A model abuse prevention program for nursing assistants. The Gerontologist, 33(1), 128-131.

40.

Pillemer

Moore

(1989). Abuse of patients in nursing homes: Findings from a survey of staff. The Gerontologist, 29, 314-320.

41.

Quinn

Zielke

(2005). Elder abuse, neglect, and exploitation: Policy issues. Clinical Geriatric Medicine, 21, 449-457.

42.

Rabiner

D. J.

Brown

O’Keeffe

(2004). Financial exploitation of older persons: Policy issues and recommendations for addressing them. Journal of Elder Abuse & Neglect, 16(1), 65-84.

43.

Rabiner

D. J.

O’Keeffe

Brown

(2004). A conceptual framework of financial exploitation of older persons. Journal of Elder Abuse & Neglect, 16(2), 53-74.

44.

Roby

Sullivan

(2000). Adult protective services laws. Journal of Elder Abuse and Neglect, 12, 17-34.

45.

Rosenblatt

Kyung-Hwan

(1996). Reporting mistreatment of older adults: The role of physicians. Journal of the American Geriatrics Society, 44, 65-70.

46.

Saveman

Astom

Bucht

Norberg

(1999). Elder abuse in residential settings in Sweden. Journal of Elder Abuse and Neglect, 10(1/2), 43-60.

47.

Shinoda-Tagawa

Leonard

Pontikas

(2004). Resident-to-resident violent incidents in nursing homes. Journal of the American Medical Association, 291, 591-598.

48.

Streiner

D. L.

Norman

G. R.

(1995). Health measurement scales: A practical guide to their development and use (2nd ed.). New York, NY: Oxford University Press.

49.

Stiles

M. M.

Koren

Walsh

(2002). Identifying elder abuse in the primary care setting. Clinical Geriatrics, 10(7), 33-41.

50.

Swagerty

Takahashi

Evans

(1999). Elder mistreatment. American Family Physician, 59, 2804-2808.

51.

Teaster

Lawrence

Cecil

(2007). Elder abuse and neglect. Aging and Health, 3(1), 115-128.

52.

U.S. Department of Health and Human Services, Administration on Aging. (2005). Long-term care ombudsman report: FY 1998 with comparisons of national data FY 1996-98: 9-9-2004. Washington DC: Author.

53.

Watts

Zimmerman

(2002). Violence against women: Global scope and magnitude. Lancet, 359, 1232-1237.

54.

Weatherall

(2001). Elder abuse: A survey of managers of residential care facilities in Wellington, New Zealand. Journal of Elder Abuse & Neglect, 13(1), 91-99.